95. Puzzle Setting, Not Goal Setting

Hello and welcome to another episode of the Operations Room, a podcast for COOs. I am Brandon Mencinga, joined by Bethany Ayers. How are things going, Bethany? It's just relentless, I think.

I don't know if that's how things are going, but that's things are feeling. I know, I'm hearing you on this loud and clear, the relentlessness of the job. Relentlessness of VC-backed businesses, which we signed up for it. We love it.

We enjoy it a good 60% of the time, maybe 70% of time. Yes, yes. This week has been a terrible week for me in this sense, which is people issues related to people not understanding the fact that these are scale ups. It's venture backed.

We need to move at pace and their personal situations or their workloads or whatever. I respect the fact the people need to like have their space, craft their worlds and so on. But there's also very much like a pacing element to this company and to venture-backed. Your point that is relentless and needs to happen so being in service about which we all are because we all signed on to this is the name of the game if you're not up to it you don't want to do it yet personal considerations don't allow you to do what you should not be signing up for this venture back journey i would say.

100%. And also, on top of just the general rollercoaster that is venture-backed businesses, we have AI and the shifting sands below us and the market changing every day, and anthropic building our products every day or somebody else's product. I know, I know. The general gist of it is, yeah, we're behind, everyone's behind, I'm behind, you're behind.

Who's not behind? I'm like, I got it. There's like another kind of CEO problem. We have like a thousand things to do.

We need to move at pace. Sands are shifting. We need get behind that to make sure that we're transforming the business. And trying to do all of it at once is like, there's only one brand and there's not three brand ends or three versions of ourselves, there is only so much you can accomplish.

But you can see I'm kind of riled up this morning as we speak. Yeah, and it's also something I've been thinking about and have talked to my team about, but I don't think it's just unique to us is not only is it like AI adoption internally, but it's this ability to shift to markets because the market's moving so quickly. It's kind of both moving quickly and not moving at all. Like there's definitely a bifurcation of the adopters and the movers.

And I feel like after open claw. There are a lot more people who are throwing themselves into AI than there were a month ago. But then you still have companies that are moving at a snail's pace who can't figure it out. And kind of like straddling these two worlds.

And it means like, there's been a lot of talk around the change of org structure and what roles are staying and what rules are going in this new world. But I don't think we've talked as much around Cadence. Operating models and speed. Some of it I thought was specific to my experience at Matomic in that we're 20 people, the goal is not to be 100 people, the goal was to be maximum 50 people.

But we've all come from bigger companies and we have a tendency to over process stuff and put in things that are unnecessary for the of people that we are. But I'm wondering if it's that or actually old way of building is let's put in the scaffolding, let's make things quite rigid so that they will hold the new bigger company. But right now, if you make things so rigid with the shifting sands, you're just going to fall and crack and break into the sand. And so what you actually need to do is figure out how to build much more flexible Processes systems ways of accessing information ways of communicating so that.

You can bend and shift as the sands move rather than break. I don't have the answer to it, but it's something I'm starting to formulate is like, how do we make our businesses more flexible? And also like right now is probably not the time to build the processes that we're familiar with. Like it's really the time to think about what's the output we're looking for and have less of a formulation on time.

I'm trying to articulate this versus like the ability to get what you need when you need it, share what you to share when you to it, rather than in a cadence that we've been used to where things were moving. Like, venture's always been fast, and we've always maybe moved on a weekly or monthly cycle, but I feel like today we almost need to work on a daily cycle. And so how do you build systems that mean that you can communicate, share ideas, move direction, change your home page.

Create a thousand different landing pages today and not in two and a half weeks. From an individual, from team point of view, there should be like a much stronger capacity to move way faster, doing a lot more things. So if you think about classic customer discovery that a product manager would do or a UX researcher, you know, that process used to be quite tedious and quite slow in terms of the processes that they go through. I think today for customer discovery, you can do that way faster.

And it's not just the product manager anymore, but it's the actual engineers themselves and the designers, everyone's involved in customer discovery in some form. And the speed by which you can, you know, the bottleneck is more of just like the customer interactions themselves than anything else required down flow from there because we can kind of AI agent orchestrate almost everything to actually get the realizations, the insights, the changes to the product roadmap, and then the actual production of those features or experiences, you now, from a kind of the coding perspective as well, I suppose.

Yeah. And so maybe it's not around flexibility, it's around collapsing time. But I think if you collapse time, that also gives you flexibility. And one of the things that is definitely a mind shift change that I see some people needing to make is, this is a hypothetical, this isn't actually something that happened, but it's just come to my mind, is by example, if we need a thousand landing pages.

I mean, we don't really need a thousand pages, but like... I feel like sometimes what we do is figure out how to build a thousand landing pages. But what we don't do is think, I'm going to need to be able to, like the operator brain, I'm not gonna only have to build 1000 pages once, I'm gonna have to do it all the time. And I'm also going to have to like change the website overnight or do that.

And so how do I actually build the whole system that then enables rapid change? In not just building a thousand landing pages, but like the entire marketing flow, for example. And I think it's as operators and COOs, it's the way our brains naturally work. And so I think maybe we need to start training people to see how our brains work, or how do we create clawed skills that are that, and everybody uses that clawed skill.

That would be the better solution, brain working in a second. I used to talk about how we need to get leadership skills into the team. And I think now how do we get system thinking and operator thinking into the team? Yeah, I can definitely see more of an operator skew going forward for anyone that's in a company because, you know, this becomes much more about judgment, decisions, system level thinking, and much less about actual execution of whatever program.

It's interesting, I was talking to one of our software developers the other day, and he was saying to me, his argument was that he feels like in a weird way, we've gone back 25 years into more sort of like weird waterfall in a way whereby if you're a product manager, You need to be doing much more descriptive, out-of-the-box, prompting specifications of what you're actually looking for with some level of clarity, going into Claude when you start moving from that point to actually create production code, and that the so-called PRD descriptions are kind of back with product managers.

I also agree that there's stuff that's like back 20 years or 25 years. I just wasn't close to product development. So it's not the one that I'm thinking about. The one that.

I do think about as a CEO and anybody who's in charge of infosec should be thinking about is how much is sitting on people's local drives in a way that hasn't happened since 2003. Because we're sucking everything out of the cloud, downloading it, feeding it back up to Claude, or whatever you're using. Or you're moving things onto your local drives, like all of it to get data closer to Claud. If people lost their laptops now, it would be such a bigger deal than it has been for the last 15 years.

You know, it's interesting. We just had this conversation the other day, but this new reality where software development, in a way, the job is done, right? And the question of in your company or in any company, you're gonna have certain developers that are on AI agents and using them wholesale to create their code. And what they're doing else we simply doing code reviews before it goes into production you have another band of your developers that there's still this kind of gray area where they're starting to take a look at using agents properly and so on but they haven't really vector themselves hard into it where some people are going hard at right now they have another bended developer where they kind of like either not interested don't want to can't see the future what have you knows that band of individual and i guess for companies right now doing.

Kind of what's sensible, I think, which is to migrate out the people that are not down with it, taking the gray zoners and trying to like enable them and push them hard into the kind of AGI-pilled environment. And then the ones that are already on it, obviously that's fantastic. But definitely there's like some level of organizational stewardship from a CEO perspective. I think that needs to occur to like just recognize this, which is software development is a dead end job.

And if we recognize that endpoint, the question now in companies is what do you do with that? How do you best enable the gray zone people, how do you best enable people that are not down with it, and how do you best kind of discontinue to support the ones that are going hard at it? Yeah. I mean, that's why we're doing quarterly training and consistently combining quarterly training with time to then build stuff and build out, you know, play with the new skills, take the time to build the rest of the infrastructure.

You know, cause every time, every quarter things have changed so much from the previous three months that there's a bunch of new things that you can take advantage of built in so that your job is better. I agree that coding... In and of itself is a dead end job, but I think technologists are not. So the understanding of how it fits together, the precision of it, how to formulate and work through and articulate the problems that need to be solved and also validate that it is going to solve the problem, that thinking isn't going away.

Yeah, and this is where product managers, designers, engineers that have more of a customer-facing orientation, there's gonna be like a weird blending of people. The competencies are still there to your point, but you're gonna see much more, I suspect, engineers being asked, forced to deal much more directly with customers or customer discovery or more of that front end of like validating actual value themselves as opposed to like a PM doing it on their behalf. So I think people who identify as being builders or who really like building are the ones who are going to.

So here's another thought for you. So this idea that if anyone's building code, whether it's a product manager, designer, engineer, or other people, you have a bunch of people, including the engineers themselves, that are committing code from Claude and nobody has looked at it. They're just like, yeah, this works, it's committed, and it's doing its job effectively. But no human being at that point really understands what they've just committed into code effectively.

So it goes into the code base. And let's say two months down the road, you're having massive stability issues and everything goes down. So because nobody actually understands the code base anymore, you start using Claude again to try to debug what is happening. Claude leads you in a direction.

That direction is based on Claude telling you where it thinks it's at. Claude is less good at doing this right now and oftentimes you go off in wrong directions effectively and what one of the senior... Developers was telling me was that he was looking at Claude's direction for the debugging process so that some of the developers were going to focus on. He looked at it and said, this is bullshit, that is totally wrong, it's actually over here somewhere.

They redirected their efforts to that other spot that he'd identified and lo and behold that's where the bug was effectively. So what he was suggesting to me was this kind of like software architecture design code review on the input side, looking at what Claude is telling you from a debugging standpoint on the flip side and being critical of that. That critical thinking is still required. But inherently, as we all are, including myself, we're all inherently lazy, right?

So we're kind of like, code reviews, that's boring, I don't want to do that, the debugging thing, Claude must be right, I'm just gonna not think so much and just allow it to tell me what to do. So what I'm wondering out loud is, like is there a space for, not just a technologist on the customer-facing side, but some technologists for like the software code base itself where there's a different set responsibilities for someone. To do some of the stuff that I think is going to be missing.

100%. That's part of what I meant by the technologists. You can't just prompt Claude and walk away and not take responsibility for Claude's output in the same way that on the business side, I wouldn't say to Claude, here are all of our numbers for the last months. Look at my emails and write me a board summary that I can send to the board without editing.

And walk away. Done. And walk away and just be like, after you've written it, just send it to the board, done. You wouldn't do that.

And so why in the world do you think, even if you spend quite a long time prompting Claude, that you would just push it without checking? Like we're not at that level of intelligence unless you have really built in a huge amount of infrastructure to contain Claude. All right, so we've gone on like a wild tangent here and all sorts of things. We've got a great topic today, which is puzzle setting, not goal setting.

We have an amazing guest for this, which is Radhika Dutt. She's the author of Radical Product Thinking, the New Mindset for Innovating Smarter, and a former chief product officer for MovePrice. So with Radhica, she was talking about, and this is a quote from her, goals and targets were designed to kill that internal drive. Where you're then focusing on external measures as if these external drivers are needed to motivate you.

And I guess the question is, do you really need to have actual targets to kind of like focus people and to motivate people? And I think she was suggesting in her puzzle framework, there was less need for the actual KRs themselves as opposed to the objective. What do you make of that? My very first reflection is, did a leadership offsite last yesterday, and we had a whole conversation about OKRs, and I entirely forgot about this conversation, and now I'm just kicking myself.

I'm like, why? Why didn't I bring this forward as an alternative of what we should be doing, because you know that I'm not a massive believer in OKR. And I think we ended up landing on not doing OKR, but the leaders of the largest Teams. Figuring out the model for their teams.

So sales is in effect, you know, running the cadences that a sales team would run and engineering or running sprints, but, and having a roadmap. But I'm wondering, now I just want to bring it in. I know I'm supposed to answer the question, but it's just like kicking myself. Like I really wish we'd had this conversation like last week.

Part of it I think depends on the size of the company, but. It feels to me a lot more true than OKRs. OKR's are just. They can be very rigid and people can end up gaming the system and caring more about getting the right key result than solving the underlying issues.

And so whether it's a puzzle, whether it is a key result, I don't think matters as much as, and by calling it a puzzle I think you reset the vibe. I'm almost feeling like it's a masculine and a feminine approach to it. So like the masculine approach is very like, okay, R and let's go into fight and like very straight. And the puzzle is...

Let's work together, let's collaborate, let's talk about this and look at it from every direction and is a bit kind of more rounded and not relaxed, but like we could put a little bit more fun in and like what we're gonna do is fundamentally solve the problem rather than drive and deliver. Yeah, so I agree with you completely. I think part of this is, I think people become a slave to the KR or they become a slave to like crafting the right KR or the become a slave to only caring about that singular number and pervert what the real intent is of the actual KR itself and don't recognize that throughout the discovery process that may change the entire game whereby that KR is entirely irrelevant or the number makes no sense for a variety of reasons and if they're not to that, open-minded enough to that.

I.e. The puzzle orientation, they're going to end up doing the wrong thing. And the manager at some point will be satisfied because they've hit their KR, but then it'll become like, well, what was the point?

Why do we bother to do this in the first place? Because it had no business impact whatsoever on the trajectory of the company, yet we hit our KR. It's the classic, I've delivered and I'm done and I am very happy as an individual or as a team, but ultimately from a business outcome standpoint, you've done jack shit and it's had no impact. That was the Cassie Young episode on first team, the top team, and you know, here marketing is saying, look, everything's green and the entire business is on fire.

So the other element that Radhika had talked about was a puzzle means you have to stay within the discomfort, not knowing the answers and embracing the not knowingness in terms of your first step. What do you make of that in the context of OKRs? I don't know if it's in the context of OKRs or more broadly the context of life, like sitting, being comfortable with the discomfort is critical to survival, like generally in the world and particularly now. Like one of the things that I tell my kids, because it's one of the things I tell myself, is when there's a setback, this really feels horrible, this is really uncomfortable.

I have survived other setbacks and what this will give me is a reliance to know that I can survive this one and the next one so I'm not afraid of discomfort because I know I can survive it. And I think if you can instill that in your teams right now, we can have a need to solve problems. It's really powerful. Although another thing just popped into my mind is We've been talking a lot about Myers-Briggs at work.

I'm not a big believer of Myers- Briggs in terms of like when you put it all together that it's realistic, but I do find the individual four pieces helpful. And the final piece is like, I can't remember, it's P and J, judging and perceiving, which is such a bizarre, like I kind of feel like they're synonyms, but people who are judging are very, very time bound and feel comfortable when they've made a decision and people that are perceiving. Feel very comfortable before they've made a decision and before they're tied down.

And I am a very natural J, and every CEO I've ever worked with is a very naturally and highly-preferenced P, and it's been such a clash. But now that I'm a CEO, I have to try and tamp down my J and leave space for the indecision and leave a space for doing the puzzle and actually remove some of my execution urge. And so I think it's a helpful structure. Yeah, no, I hear on this because I feel like I ride the lightning of this as well in some ways because just with the past week that I've had, I've had to make decisions at pace and I get pretty excited and jazzed up to make decisions like I'm a natural born decision maker.

And sometimes it catches me out because I'm making calls either too fast with not enough information, lack of context, all sorts of things. So sometimes it can be an asset, but sometimes it's gonna hurt you. And we're just having a call with one of our external recruiters. And I was telling the recruiter alongside of our internal staff, I'm like, we're doing this, you're not doing that.

And then at some point he piped up and said, hey, Brandon, like we actually have some skills in this particular area that they may not be aware of. And this is what the skills look like. I'm, like, oh, fine, love that. Let me pull back what I just decided.

We're doing that now. I'm getting out of myself slightly from a decision standpoint, I would say. So I think having this conversation helps us be aware of when you do push for a decision and when you maybe push for micro decisions but leave the full solution to unfold. Yeah, yeah, being open to it.

That makes complete sense. Last question, Radhika talked about the three O's from her point of view, which is the key bits are observation, the open questions, and the objective. So when you think about the 3 O's as kind of like her puzzle way of going forward as opposed to OKRs, how much do you love the observation, open questions and objective? And in particular, on the observation side, do you have any thoughts on that one?

I think it just makes a lot of sense and I think you can tell that she's a product person who is now thinking through and standardizing the process that she has created and figuring out what it is that you do. So you observe what's happening, get really clear on that, ask the questions that aren't yes no questions. There's a bit of a debate as to whether or not open questions actually matter. I think it's more around, like, a philosophy of the mindset, because...

You can ask a very open question to somebody who doesn't care and they'll still give you a closed answer and you can ask the closed question to someone who cares a lot and they're going to give you an open answer. So I think it's more inviting a little bit more of that P versus J. So, you know, like inviting the conversation, inviting not needing an answer and surfacing everything that you need to know in service of an objective. You're absolutely right.

I think Radhika, as a former product person, she's taken effectively the idea of customer discovery and applying it to OKRs and re-vectoring it as puzzles. So she's very much taking her background, I think, in kind of this new thesis that she's presenting. It's not really new, is it? It's just kind of product management, customer discovery repackaged into the OKR world of running companies.

And a discovery that it works, yeah. Yeah, no, exactly. The interesting bit is this word observation that she's stuck in here. I don't know if you recall our talk with Keith Wallington as to the CEO's job.

He said one of the five jobs of the CEO is observation. So it's just interesting, word-wise at least, that Keith saw that from a CEO perspective to spend time on that. And she's kind of saying the same thing. So why don't we park it here and get into our conversation with Radhika.

I am not the biggest fan of OKRs, but for me, it feels like the least worst option. And so when you approached us and said you had an alternative to OKR in a different approach, I just got so excited. And I think we ended up speaking for about an hour, and I was just going, what do we do? How do we do it?

And the way that I summarized it is looking at everything as puzzles. And I love puzzles. So, I do the crossword puzzle, wordle, connections. All I only do three of the LinkedIn puzzles every single day.

And so for me, the idea of looking at work as a puzzle and introducing more play, particularly when things are so stressful is a stroke of genius. So now I'm going to let you talk and I'm just going to say, why don't you introduce us to this puzzle concept and puzzle framework? You know, but I love your preamble to all of this. This is what I'd been seeing for a really long time in my career.

And I couldn't articulate quite why goals and OKRs weren't working. What I was feeling was we just have so much passion and internal drive. And it's almost like goals and targets were designed to kill that internal drive where you're then focusing on external measures, but not just measures, external drivers as if you don't have this internal drive that these external drivers are needed to motivate you. It was fascinating for me for a long time to see the kind of twisted and perverse effects that goals and targets had.

This puzzle concept came to me because I just instinctively started working with a team in this way. So let me share this example. I was working with this Greek company, I still am. The maritime industry.

The way this started was because the CEO had brought me in. They used goals and OKRs and sales had stalled in 2023. When I came in, the team had been trying to go after this market and they were hitting some OKR's, not hitting others. But it was a brilliant group of people, super smart people.

All were data scientists, mathematicians, applied mathematicians. Like really fascinating people. And clearly, there was passion when I was talking to them. And yet, when I actually was seeing the work, I wasn't seeing that passion translated into that work.

This was when we started looking at it differently. Instead of focusing on goals, targets, and looking at, okay, is this feature going to move this okay or not, we just started with this blank slate saying, okay wait, let's just look at what is the problem we're solving? And we started thinking about it as a puzzle. You know, I just wanted to know, how are things going?

What is it that customers are really experiencing? What is working? What is not? And so we started defining the puzzle.

So this was a maritime platform, a data analytics platform. And the puzzle was, it's fascinating that only tech savvy brokers in the maritime industry are using this platform. So fascinating. Why only tech savvy people?

Why only brokers? Why aren't people who own shipping vessels, or who own cargoes, why aren't they using this platform? Why just brokers? And so these are questions that we didn't have answers to, right?

And when I started asking these questions, people are like, yeah, you're right. We don't know this. A puzzle mindset is different because goals and targets come at it from, we know the answer. Just achieve these numbers, that gets you what you need.

That means progress. Whereas a puzzle means you have to stay with the discomfort that we don't know the answers. We're going to have to figure this out. That embracing the not knowing is the first step of a puzzle that completely shifted the mindset.

So we started talking to people who are both customers, people who weren't customers, to figure out what's really going on. Why aren't these people using it? And it led us to questions like, how do people who are shipping vessel owners use, what do they do in their day? How do they work?

You know, what is their mental model? How is it different from how brokers work? So we started asking all these questions and we learned so much. We were observing what was actually happening, not just looking at data and not looking at shortcuts.

We were absorbing this puzzle. It's like looking at a Rubik's Cube and really moving things around and saying, how does this thing work? And now, now that we had this observation, we were looking at this puzzle, we had a clearer definition. Now we could attempt this puzzle.

And so I'll pause there for a moment, but I think there's lots to unpack even there. And then I can talk about how we solve the puzzle. One of the things that struck me immediately was, and you said it very in passing, you were really observing things rather than just looking at the data and short-cutting, and I was like, oh yeah, I shortcut. I look at the date, I make a close enough guess, and then I say, now we need to build X amount of pipeline, or now we'll need to ship these three things and move immediately into the goal setting side and skip the discomfort.

But there's also an element... Where I'm wondering, you knew the problem. And so like, the part that we're missing is the beginning of what were the OKRs, or what was the OK that didn't work versus figuring out the problem to solve, because presumably they were trying to solve the problem through the OK. Yeah, we were.

We wanted better user engagement. That was one of the OKRs, better user engagement from all these different types of users, right? We wanted to have more customers who are not just brokers. We want to expand to this other user base.

We wanted to expand beyond just a small group of brokers. And so this is when we started looking at the data to see, well, who is using this? That was when we could tell, okay, it's mostly tech-savvy brokers using this. And yeah, very few other people are using this, right?

So yes, you're right that we did have OKRs saying we should do this, this, and this. OKR's don't tell you what is it that's not working, right. And so we were hitting some of these numbers. We were increasing some usage, but not enough.

Like it wasn't truly a radical shift. And do you know what you were saying about... How we want to jump to the quick fixes, that was the temptation at this company too. When I first joined, they were looking at the data and they were saying, aha, we know what we need to do.

All the customers are complaining that filtering on this platform to get to the data you want, filtering is really hard. And so we need to make filtering easier. Let's work on the UX to be able to change filtering. And even there, right, we wanted to jump to the quick fixes.

Yes, let's work on filtering UX changes, but don't make radical shifts because that's going to take a long time for us to build. Let's do something that can increase our metrics and let's try something small first. So this is exactly what you're talking about. And that was the trap that we were falling into until we said, but hang on, let us first observe what is the puzzle.

Is filtering truly why people who are in different roles, not brokers? Is filtering truly the reason that cargo owners or ship owners are not using this product? We didn't think, like I didn't, think so. It wasn't obvious to me that filtering was the main reason.

The people who are using the product were complaining about filtering. The unknown unknowns were what we didn't know. The people were not using it, there was no data on them. The way them suddenly starting to think about it is maybe they work together or you know so you can use OKRs in terms of what are the things to identify what the not the solution but what the outcome is and like what you're trying to get to but so there was a very clear business problem which is you need to have more users and users across different bases and measuring the success of that is important.

But what ends up happening I think a lot in businesses is you set out the OKRs, you set out the numbers, and then people immediately game the system or it's like it's a quarter. How do we do it as fast as possible? How do you do it efficiently? How do cut corners so it looks like we're doing it, but really all we're doing is measuring it slightly differently than we were before and we're not actually getting to the true unlock.

And so it's the two together. You need to identify what what needs to change, but then not make reaching it the goal. What you need to make is unlocking it and truly understanding the problem, the goal? Exactly.

So, when you set a goal or a target, the incentive you create for everyone is to show you, tada, look, I'm a high performer, I've hit whatever numbers you wanted. And so, the incentives you create by default is to get everyone to tell you things are going great and I have the numbers to prove it. And you know, in fact, Intel's legendary CEO, Andy Grove, he wrote in his book, Leaders are the last to know. And this is precisely why, because when you set a goal or a target, everyone is there to show you.

And it's not even malicious, right? It's not that they're necessarily gaming numbers on purpose. Even subconsciously, what happens is you set a target and people want to look at the positive numbers and present to you those positive numbers to say, tada, look, things are going well. And what you really want people to do is to play detective, look at the bad numbers too.

And say, huh, what just happened there? You know, is there an opportunity? Is that a puzzle? What went wrong?

And what can we do differently based on that, right? And that's what we miss when we set targets. And so what you're alluding to is how do you use this in an OKR format? So what I do with puzzle setting is I call it the three O's of puzzle setting.

So the first O is the observation. So the observation in this case, right, was. That only tech savvy brokers are using this product, none of the others are. We want to address the tech-averse people, we want to address ship owners, the cargo owners.

The second O is the open questions. These are things that you genuinely don't have answers to. This is where we don't know how these other people think in this business. How do they think?

How is their mental model different? And so that's the second O. And the third O The objective, and by the way, the objective is named very deliberately to be similar to OKR's objective, right? Because that objective is where you summarize this puzzle.

And the summary of this particular puzzle was, how do we solve this puzzle to get back to the growth trajectory we were on by expanding to all of these different sorts of users who are not just the tech-savvy brokers? And so if you're using OKRs instead of just having arbitrary you know, objectives like grow sales or increase user engagement or whatever that is, right? Set the whole puzzle in this very clear way so your team knows exactly what they're going after in terms of puzzle solving next.

And are those O's in order? Absolutely. So you observe, I can't remember the middle one, and then objective. The open questions.

Open questions, so that's part, so you observe ask open questions, figure out what's going on, set your objectives, and then solve the problem. Exactly. Let's take a sales example, right? So the sales example might be, you know, instead of just setting the target as grow to X million at the end of this year, the puzzle I might set is, we do want to get to X Million, but here's the puzzle I see.

We grew at a certain rate in the last three years and growth has stalled, right? That's the observation. The open questions are, I don't know what just happened. Is it that maybe?

Something is fundamentally shifted because of AI that is not obvious, so even though our market isn't related, there's still something that's affecting our business. Maybe it's that we were targeting early adopters really well, but the mass market, this message is not resonating for them, right? There might be other such open questions that we genuinely don't know what's happening. And then the objective is the summary of this puzzle.

How do we get back to that growth trajectory by figuring this out. The way that you're talking about it, it strikes me that these are the conversations that we have as a leadership team, and it's a lot of how to frame the strategy and understanding the full context, but we don't always then explain it to the rest of the team. So is part of this around a way of explaining and getting the conversations out of the leadership and into the rest to the organization? Very much so.

And And it creates a level of alignment that is so different from how we traditionally communicate strategy. What I often see in companies is, there is discussion that happens at a leadership level, but what gets handed down to others across the organization is a very fluffy two-page strategy document that says, we're going to focus on our core, we are going to be customer-focused, and it doesn't mean all that much for people. Or it's a particular initiative that is listed, instead this defines sort of puzzles that we're setting out to solve.

And each puzzle, there might be a sales component to it. The same sales puzzle might have a product component to, it because maybe the product was resonating for the early adopter, but it's not resonating for the mass market. Like there's a lot that might be even cross-functional that needs to be figured out. So that's very much around the puzzle setting piece.

And I have some questions on how big are the teams, how many puzzles can you solve at one time, like those COO practical questions. And I'm not sure if that's part of the, if we're moving into the puzzle solving piece there or if that still in the setting. The puzzle setting part, right? I think it depends on the level that you're at.

As CEO, the sales puzzle that I defined, where it has a sales component, where is our message resonating? Why aren't our deals closing? There might be a piece of that puzzle for the sales team to figure out. There might a product piece for the product team to go figure out, so at a CEO or COO level, you might be setting a bigger puzzle, right, and maybe as a head of product, there might be.

A smaller version of that puzzle. And in this particular case, the puzzle was really at a product level, which was why aren't we growing beyond this group, right? And then once we delved into that, there were smaller puzzles. So as we get into puzzle solving, I'll share with you some examples of how solving one layer of the puzzle, kind of like the Rubik's cube, you solve one layer and the next layer becomes the next thing that you're solving and you keep going.

And, you know, this question that you asked, what it reminds me of was Andy Grove's book. You know, he talks about only the paranoid survive. He talks about how you have to keep experimenting, learning, adapting all the time, right? He doesn't even talk about OKRs, even though he invented OKR, and it came from Peter Drucker's Management by Objectives, but he inventedOKRs.

He never once mentioned OKR in only the paranoid survive, it was all about experimentation, learning, adaptation. And with each team and at different levels, you get into different pieces of the puzzle. So is this purely just a framing exercise from top to bottom, because it's kind of the same thing, right? Because whether you call them OKRs or you call the puzzles or you called them whatever, I recognize the framing matters quite a bit and also being bound by rules matters quite a bit in terms of giving freedom for experimentation and so on.

But ultimately at the end of any OKR, if they're done well, which they're usually not, but the team, the cross-functional team will have a set of hypothesis or initial . Things they want to put in play to test out as hypothesis to see if it's going to move the needle related to their KR, I suppose. So inherently, those initiatives are puzzles because they have a hypothesis. They're like, yeah, we don't quite know, but we're going to like, this is our first thing we're gonna salvo-wise get out there to see if it actually moves the needle or not.

So I guess my question in summary is the puzzle, part of it is part of generally what happens in OKRs at some level. So it's more just like a repackaging and reframing of like what we're doing here as opposed to something more fundamental. You know, what is funny for me is this is actually what I hear from so many OKR experts that, you know, if you just do OKRs right, then you don't run into any of these problems with OKR. The reason OKR aren't working for you is because you haven't done OKR right.

This is sort of the refrain that you keep hearing. Here's my challenge to that refrain. It feels like the emperor has no clothes and someone has to call it out because It's not that if you do it right, and I'll give you examples even from John Doar's book, who is the one who evangelized OKRs. Even from his book, I can give you an example of OKR's that are supposedly done right and the ramifications that come from it.

OKR create a different mindset. What research has shown about OKR and specifically about setting targets is targets and goals, they work well. When you're doing a repetitive task, when there is one obvious right answer. So research has shown that if you're doing something like stuffing envelopes, if you are doing crunches in a gym, great, set targets, it works brilliantly.

If on the other hand, you're doing something that's more like a puzzle, that's akin to something complex where there's no single right answer, that is where goals and targets do not work. That the instruction, do your best, actually works better. Than setting a goal or a target. So why is that, right?

And it begs the question, how did we even adopt goal setting as such a fundamental truth in business? And the answer to that, I did research on this because I found this fascinating. Why haven't we used puzzles? Because that seems more obvious.

Where did we come up with goal setting? It came from 1940s from Peter Drucker. It was revolutionary at the time. And why did Peter Drucker then come up with this goal-based approach and management by objectives?

What problem was he solving? The problem was that he was working with General Motors at the time where the problem they had was motivating people on an assembly line who are doing repetitive tasks. Now it all falls into place. Of course, on this assembly line in the 1940s, there was no automation.

It's all repetitive tasks and so goals worked really well, exactly as research shows. And now we use the same approach, even in a manufacturing setting. Look at Boeing. You set production targets and you see the kind of quality issues you have.

So it's not a matter of just setting the right OKRs. Sorry, I didn't mean to derail your next step here, but I'm now very curious about the second step. We've set our scene and now we're going into it. Give me the low down here.

So puzzle solving is where, instead of proving that I've hit the target or the number, the incentive is different. We're asking three questions in puzzle solving. Until this point, we've looked at the Rubik's Cube and how it works, and puzzle solving is when we take our first attempt at it and we ask three questions, which is how well did this work? And notice the difference, right?

How well did it work invites the good and the bad. It's not like a target. I'm not asking a binary question. Did you or didn't you hit the target?

I want to know how well it worked. And so the answer, let's say in the sales question, was, okay, we tried this new messaging. How well did it work? The messaging worked well.

We got all these meetings with decision makers, but we're not closing the deals. So that's how well that worked. The next question. What did we learn from it?

So this is where I say to teams, don't just spit out statistics and data at me, right? Data and stats are super important, but I want you to take the time to figure out what is it telling you? What is actually happening? And so in the sales example, it would be that the message is resonating, we're getting meetings.

The problem, the reason why we're not closing the deal is that in addition to this decision maker in the mass market, it turns out there's a new group. That we traditionally had never talked to, and this group is holding up the sale. So that's what have we learned. So now comes the third question, which is based on how well it worked and what you learned, what will you try next?

And this is where I say to teams, you know, imagine I give you a magic wand, what would you ask for? So let's start from there where you really rethink things and then we'll pare it down to what we can do. What would you asked for? And in this piece, right, to solve the sales puzzle and what we tried next, One example would be.

We're going to try creating material that the decision maker can easily forward to this new group. We're also going to reach out to them directly. We traditionally have never spoken to them. Here's how we might reach out directly.

We're gonna do webinars. And now that opens up the next layer of the puzzle. We're trying these things and see how well did it work. What have we learned?

What will we try next? It feels like this is how I operate, but I just have never put into a framework. I don't know about you, Brandon. Like it seems like I have these kinds of conversations on a fairly regular basis.

Yeah, for sure. Yeah. And I guess I'm just wondering, like, what is the process then? Yeah, and so isn't that fascinating?

Like, I agree with you. Intuitively, this is how we work. This is how you solve puzzles. And yet, right, this not how I actually see teams operate.

What this gives you is a scaffolding so that what you find is people at different levels, they have different levels of knowledge, skills, and experience. And this scaffolding helps you figure out where is someone at? There have been times where I was working with someone who was a product manager, someone who came from Amazon. She was fantastic at optimizing for numbers and she would often present numbers that looked fantastic.

And we started using this framework and I was asking these deeper questions. How well did it work? What have you learned? And I would start to question some of the numbers in the, what have we learned?

Like, I don't just want to know that 48% of people are using this. Tell me the details around it. How did you get to the 48%? Is it that they're landing on it or is it that they're going back to it?

Like, what is actually happening? I want the details. When I started asking these questions, I would realize how well does someone actually understand what is happening. It's a different level of puzzle solving, right?

They're not just telling me numbers to make things look good. We're actually solving the puzzle. And then I want to know, what are you going to try next? All of this tells me exactly where someone is in terms of knowledge, skills, and experience.

And it helps me scale leadership, or rather, it helps me scale my own team's efforts. Like, based on who is good at solving what level of puzzles, I know how much to delegate. So that Reminds me of another guest that we had on Jennifer Sonderberg from Board Intelligence, the episode on how to build a company that's smarter than you are. Board Intelligence are a company to build good board reports, like for multinational companies for the most part, where board reports are big and dense and don't share the information that people need to know to actually govern businesses.

What they came up with was a questioning framework. And the power of the question, and using that as a way of embedding critical thinking in an organization. And it seems like what you're doing is also solving the puzzle of how do you embed critical thinking in an organization? And the thing that has definitely been my experience with OKRs and goal setting is that it encourages superficial thinking and it encourages moving fast for a result.

Rather than the critical thinking. And so you end up with a leadership team that might be thinking critically and everybody else is just executing and cutting corners and moving fast in order to hit numbers or not believing in those numbers and those objectives. And so then just like not doing it anyhow or looking like they are, but knowing that it's unachievable. So I can really see that you need a business plan, you need KPIs, you need.

North Star, you need to know where you're going to and some measurements. But OKRs might encourage a lack of critical thinking. Or you can have a lot of critical thinking and use OKR as a tool, but you'd have to be embedding your critical thinking somehow. And if that's solving puzzles or asking questions, that's really what you're trying to achieve.

Exactly. So what you often hear in the corporate world are all these trite phrases, right? Like, we have to embrace failure, we've to encourage people to experiment and we have to make it okay to fail. And, you know, when you think about it, it just seems so contrived, right, because you said you love to solve Wordle and all these different puzzles.

Let's think about, you know, when you last solved Wordle probably earlier today, when you tried your second attempt, if you didn't get it on your second attempt, did you artificially go through this thought process of, I'm going to embrace failure and it is okay to fail and I'm gonna try this next experiment? No, right? We just sort of instinctively look at something that didn't work and we go, oh, here's what I learned from it, here is what I'm to do and adapt next, right? And I think you're puzzle solving.

Strength is why this feels so intuitive to you and you bring it into business. What we want to do is really to spread this passion, but also this drive, this internal drive that people have. We just want to enhance it so that they can use the scaffolding to solve puzzles in an effective way so that we're all moving forward in the same direction. That I also like about it being called a puzzle is that it turns it into play and it removes some of the stress and I feel like right now there is just so much stress in the world.

We have macro stress like is there going to be war in Europe? Is there going to be a war in Canada? The planet's still burning up but nobody cares about it right now because we have more pressing issues and so there's just this like of control sense of dread. And then you add AI.

Which is a runaway technology. Nobody knows what's gonna happen to their jobs. We did the internet, and I think everybody's much more positive because we didn't realize what was gonna happen. And then social media, and now there's like more fear of AI, partially because of our previous experiences, partially because Terminator.

With a backdrop of just everything is scary. And then what is not going to work in organizations right now. Is people are so afraid. It's very hard to then get inspired, find flow, play, open up your mind for creativity.

And yet now is the time where we need to do it more than ever. And so how do we as leaders introduce and lighten things to actually get better results out of our people? And I feel like doing something like this and calling, and that's I think why I love it as a puzzle so much, is it just removes a layer of stress. And I know sometimes they're like, no, no, we need the stress to perform, but we have so much stress right now that we don't.

We need play to perform. I so agree. With you. And I've tested this in workshops, in talks.

I've tested with live audiences where I say, just think about the following two questions I'm going to ask you and tell me how you feel at the end of the two questions, at the end of each question. So just think about how you feel and the contrast. And I'm gonna ask you those two questions. What goals do you want to achieve for your company this year?

And the second question is what puzzles do you want to solve for company this year. And the answers I get are vastly different. And I'd love your take on it. So I wasn't thinking about the actual goals, but when you said, what goals do you want to achieve?

I get a bit of a hardness in my stomach, a deflating of my soul, and I'm like, oh, and it feels like really hard work. And then when you say, what puzzles do you wanna solve? Because I love puzzles so much, I'm, like, ooh, I physically lean forward, I have a lightness and a surge of energy. And it's like, yes, let me go and have some fun.

And I really could feel those physical reactions to those words. There's one. I wanted to mention about AI because you brought that up. One of the big things that I see as dangerous, this combination of using AI and being so OKR driven, it's a super dangerous combination.

Why do I say that? Because AI is going to be fantastic at optimizing for metrics, right? The better it gets at whether it's content creation, UX, coding, etc. It's going to be fantastic and optimizing for numbers, but it's not necessarily good for long-term business results.

And one example that I want to give you is even without the use of AI, but now imagine AI used for this purpose. Let's think about dating apps. The entire dating industry has been in a slump. Why?

Because dating apps optimize for user engagement. Tinder came out with swipe left, swipe right. Everyone copied it because user engagement was through the roof. And what that led to, It gamified intimacy, and it made human interactions more toxic.

And the end result was people feeling dating fatigue, burnout, FOMO, deleting dating apps. And so... Optimizing for metrics didn't lead to long-term good results for the dating industry. Bumble last year laid off 30% of all of its staff.

And now imagine with AI, we can accelerate all of that hugely, optimize for all sorts of metrics, but we're not going to create neither better products nor better things for society. Want to ponder? I like it, then I guess it all boils down to watch out what you wish for because you might just get it. And so you get all, you focus on these metrics and they end up sowing the seeds of destruction.

And which is why we need to think of puzzles instead, so that you're not just optimizing for metrics. Like the puzzle should have been, how do we make dating a really good experience for people while also monetizing? If our listeners can only take one thing away from today, what is it? As much as people say OKRs will work for you if you just do it right, that has never been my experience.

There are fundamental issues with it. And what we need is a mindset shift from goal setting to one of puzzle setting and puzzle solving. And this example that I gave you of this maritime company I've been working with, sales had stalled in 2023. We used puzzle setting in puzzle solving, we doubled sales in 2024.

Again in 2025. And we've reduced customer churn from 26% to 4%. You can drive real business results with puzzle setting, puzzle solving. You don't need goals for that.

Radhika, thank you for joining us in the operations room. I love puzzles. I will rebrand my OKRs as a puzzle. I read your book.

You have a book, don't you? Well, I'm working on my second book, actually, so it's going to be about why goals and OKRs don't work. But in the meanwhile, people can download the free toolkit and they can tell me about their experiences using this framework. And so feel free to reach out to me on LinkedIn to share those experiences.

And it might just make it into the book as a case study. Love it. Radhika, thank you so much. If you like what you hear, please subscribe or leave us a message and we will see you next week.